Evaluation of a device for monitoring sterility of injectable fluids.

An evaluation of a sterility-testing device (Addi-Chek) that uses total-sample membrane filtration is presented. Parenteral nutrition solutions were deliberately contaminated at low levels with 16 bacterial and four fungal strains. Some contaminated solutions were stored at 20 degrees C or 4 degrees C for up to 24 hours before testing; the rest were tested immediately. Additionally, one liter of pharmacy-prepared infusion fluid was tested each day for 100 days. Broth (added to the Addi-Chek unit after filtration to permit microbial growth) was cultured when it became turbid or after a 10-day incubation. Fifty-nine deliberate-inoculation tests were done. The organism was recovered from 56 of these; the other three tests may have been negative because no organisms were present in the aliquot used for inoculation or because of experimental error. The number of organisms used for inoculation varied from 1 to 80; 86% of the tests used 25 organisms or less. Storage for up to 24 hours at 20 degrees C and 4 degrees C had no effect on the results of sterility testing. In testing 100 units of pharmacy-prepared fluid (not deliberately contaminated), one unit (1%) was found to be contaminated. Pharmacy personnel correctly identified positive and negative Addi-Chek units based on visual turbidity of the broth. It is concluded that Addi-Chek is easy to use and effective in detecting low-level contamination in intravenous fluids.